hnp (version 1.2-7)

orange: Orange tissue-culture experiment data

Description

Number of embryos of orange variety Caipira produced with different sugar types.

Usage

data("orange")

Format

A data frame with 150 observations on the following 4 variables.

block

a factor with levels 1 2 3 4 5

sugar

a factor with levels Maltose Glucose Lactose Galactose Sucrose Glycerol

dose

a numeric vector

embryos

a numeric vector

Details

To study the effect of six sugars (maltose, glucose, galactose, lactose, sucrose and glycerol) on the stimulation of somatic embryos from callus cultures, the number of embryos after approximately four weeks was observed. The experiment was set up in a completely randomized block design with five blocks and the six sugars at dose levels of 18, 37, 75, 110 and 150 mM for the first five and 6, 12, 24, 36 and 50 mM for glycerol, see Tomaz (2001). The main interest was in the dose-response relationship.

References

Moral, R. A., Hinde, J. and Demétrio, C. G. B. (2017) Half-normal plots and overdispersed models in R: the hnp package. Journal of Statistical Software 81(10):1-23.

Examples

data(orange)

require(gamlss)
fit_nbI <- gamlss(embryos ~ block + poly(dose, 2) * sugar, 
                  family=NBII(), data=orange)
                  
d.fun <- function(obj) resid(obj)

s.fun <- function(n, obj) {
  mu <- obj$mu.fv
  sigma <- obj$sigma.fv
  rNBII(n, mu, sigma)
}

f.fun <- function(y.) {
  gamlss(y. ~ block + poly(dose, 2) * sugar, family=NBII(), data=orange)
}

if (FALSE) {
hnp(fit_nbI, newclass=TRUE, diagfun=d.fun, simfun=s.fun, fitfun=f.fun)
}

fit_pred <- gamlss(embryos ~ poly(dose, 2) * sugar, family=NBII(), data=orange)
orange.pred <- rbind(expand.grid(sugar=levels(orange$sugar)[-6], dose=18:150),
                     expand.grid(sugar="Glycerol", dose=6:50))
orange.pred$pred <- predict(fit_pred, newdata=orange.pred, type="response")
require(latticeExtra)
trellis.par.set(strip.background=list(col="lightgrey"))
xyplot(embryos ~ dose | sugar, scales=list(relation="free"), layout=c(3,2),
       data=orange, col=1, xlab="Dose levels", ylab="Number of embryos") +
  as.layer(xyplot(pred ~ dose | sugar, type="l", col=1, data=orange.pred))